Process and machine for automated agglutination assays with image automated evaluation

Abstract

The machine is configured to perform an automated rapid plasma reagent (RPR) agglutination test or other agglutination test. The machine includes a sample rack with multiple sample locations thereon and a reagent rack for storing of reagent. A shaker assembly supports at least one microtiter plate or other well supporting structure thereon with a plurality of wells in the plate. An automated pipette accesses samples and reagent and deposits them within wells of the microtiter plate. The shaker assembly shakes multiple samples within the wells of the microtiter plate. Finally, a camera photographs the wells of the plate, preferably from above with a light source below and the plate at least partially transparent. The image is then analyzed in an automated fashion to determine whether a ring of contrast material has remained smooth indicative of a non-reactive sample or has agglutinated/clumped together indicative of a reactive sample.

What is claimed is: 
     
       1. A method for determining whether a sample is reactive or non-reactive when combined with a rapid plasma antigen reagent in an agglutination test, the method including the steps of:
 combining the sample with the rapid plasma antigen reagent, the rapid plasma antigen reagent including contrasting visibility particles therein; 
 locating the sample and reagent in liquid form on a substrate, the combined sample and reagent having a circular shape on the substrate; 
 rotating the substrate to cause the contrasting visibility particles to form at least a portion of a ring within the circular shape; and 
 evaluating the ring to determine if the sample is reactive or non-reactive, wherein a ring comprising dispersed clumps of particles is indicative of a reactive sample and a substantially uniform ring without clumping is indicative of a non-reactive sample. 
 
     
     
       2. The method of  claim 1  wherein said locating steps includes the sample and the rapid plasma antigen reagent in a common contained space. 
     
     
       3. The method of  claim 2  wherein said common contained space being one of a plurality of wells in a microtiter plate. 
     
     
       4. The method of  claim 3  wherein the wells have a concave bottom. 
     
     
       5. The method of  claim 2  wherein said locating step includes sequentially placing the sample and the rapid plasma antigen reagent into the contained space. 
     
     
       6. The method of  claim 1  wherein said rotating step includes rotating by an eccentric mass on a rotating output shaft of a motor, and with the motor coupled at least indirectly to the substrate comprising a contained space where the sample and the rapid plasma antigen reagent are located. 
     
     
       7. The method of  claim 6  wherein the radius of the eccentric mass away from the output shaft of the motor is between about 5 millimeters and 15 millimeters horizontally, so that the amplitude of the rotating is between about 5 millimeters and 15 millimeters and wherein the contained space has a size of up to about 15 millimeters in diameter. 
     
     
       8. The method of  claim 7  wherein said rotating step includes the eccentric mass located about 10 millimeters horizontally away from the output shaft of the motor and rotating occurs at about 100 revolutions per minute. 
     
     
       9. The method of  claim 1  wherein said evaluating step includes visual evaluation of the sample after said rotating step by a human eye. 
     
     
       10. The method of  claim 1  wherein said evaluating step includes automated evaluation by photographing the sample and rapid plasma antigen reagent after said combining step and said rotating step; processing an image produced by said photographing step to determine if the sample is reactive or non-reactive. 
     
     
       11. The method of  claim 10  wherein processing said image includes production of a digital image, the digital image processed by having a curve fitted to the ring within the digital image, multiple radial lines substantially perpendicular to the curve evaluated for darkness of pixels within the digital image which fall along the radial lines, quantifying variation of darkness of pixels along the radial lines, averaging of darkness variations of said quantifying step with darkness variation values of adjacent radial lines, calculating a difference between darkness variation of said quantifying step for each radial line and an average of darkness variation values of adjacent radial lines, and correlating this difference of said calculating step with clumpiness of the sample. 
     
     
       12. The method of  claim 11  wherein the differences of said calculating step are each correlated into separate bins of data having similar values, with radial lines having lowest difference grouped together and radial lines having highest difference grouped together, and radial lines having similar intermediate differences grouped together, and counting the number of lines associated with each of the bins. 
     
     
       13. The method of  claim 11  wherein said multiple radial lines substantially perpendicular to said curve include at least 360 radial lines which are each at least 20 pixels long. 
     
     
       14. The method of  claim 11  wherein regions of the digital image outside of the multiple radial lines substantially perpendicular to the curve are evaluated similarly to the radial lines which intersect the curve, to identify regions of dumpiness outside of the curve fitted to the ring within the digital image. 
     
     
       15. The method of  claim 11  including the further step of comparing known reactive and non-reactive sample data sets processed through said evaluating step with a new digital image processed through said evaluating step to correlate the new digital image to those known to be reactive or non-reactive and to similarly score the new digital image as representative of a correspondingly reactive or non-reactive result. 
     
     
       16. The method of  claim 15  wherein said comparing step includes processing the known data set by production of a digital image, the digital image processed by having a curve fitted to the ring of dark high visibility particles within the digital image, multiple radial lines substantially perpendicular to the curve evaluated for darkness of pixels within the digital image which fall along the radial lines, quantifying variation of darkness of pixels along the radial lines, averaging of darkness variations of said quantifying step with darkness variation values of adjacent radial lines, calculating a difference between darkness variation of said quantifying step for each radial line and an average of darkness variation values of adjacent radial lines, and correlating this difference of said calculating step with dumpiness of the sample, and further grouping the differences between radial line darkness variation and an average of darkness variation of adjacent lines into separate bins of data having similar values, including evaluation of regions of the digital image outside of the curve fit to the digital image data, and with the radial lines substantially perpendicular to the curve including at least 360 radial lines and similarly processing the new image with similar results in evaluating the new image scored similarly to produce a similar conclusion with those of the known data set. 
     
     
       17. The method of  claim 16  including the further step of evaluating the standard deviation of the darkness variation values quantified relative to darkness variation values of other adjacent radial lines. 
     
     
       18. The method of  claim 17  including the further step of evaluating area under a curve plotting the darkness variation values quantified relative to other adjacent radial lines. 
     
     
       19. The method of  claim 18  including the further step of providing a plurality of separate bins into which similar data is correlated for each of the radial lines. 
     
     
       20. The method of  claim 19  including the further step of providing at least six separate ones of the bins and evaluating the sum of the last three of the at least six bins representative of the highest difference in darkness variation of each radial line relative to an average of adjacent radial lines.